The B2B buyer shift to AI

B2B software buyers are increasingly consulting AI chatbots before speaking to sales teams, reading review sites, or even opening Google. What Metricus found through our AI visibility report work with SaaS companies is that this shift has created a new category of competitive risk: brands with strong products and strong Google rankings that are completely invisible in AI-generated recommendations.

The numbers are significant. 37% of B2B buyers now use AI as their first research channel. For enterprise software decisions, the research phase often starts with a question like “what are the best CRM tools for mid-market companies?” asked to ChatGPT or Perplexity. The brands that appear in that answer enter the consideration set. The rest do not.

Why market share does not equal AI visibility

What we found when auditing SaaS brands across categories is that market leadership does not guarantee AI visibility. A category leader with 30% market share can score below 60% on AI visibility while a smaller competitor with strong third-party coverage scores above 80%. AI models do not know your market share. They know what sources say about you, how consistently that information appears across sources, and whether your language matches what buyers ask.

The CRM test: what we found

We tested AI visibility for CRM software — one of the most competitive B2B categories. What we found was revealing: well-known CRM brands that dominate Google search results were sometimes absent from AI recommendations for specific buyer segments. A brand that appeared for “best CRM for enterprise” was invisible for “best CRM for real estate agents.” The product had not changed. The buyer’s language had, and the brand’s content did not cover that vocabulary.

AI visibility varies dramatically by SaaS category

What we found across multiple SaaS categories is that AI visibility patterns differ significantly. Categories with clear market leaders and established review coverage (like CRM) tend to produce consistent AI recommendations for the top 2–3 brands. Emerging categories (like AI-powered writing tools) show more volatile AI recommendations because there is less consensus in the training data. Niche categories (like lab information management systems) show strong visibility for specialists and near-zero visibility for generalists.

Segment-level blind spots

What we found across our SaaS audits is that the most dangerous form of AI invisibility is segment-specific. A brand can score well overall but be completely absent for specific buyer segments. In the CRM category, a brand visible for “enterprise CRM” questions was invisible for “CRM for nonprofits,” “CRM for real estate,” and “CRM for startups.” Each of these segments represents real buying intent, and in each case, competitors with targeted content for those segments captured the recommendation.

The segment-level analysis is often more actionable than the overall score. Knowing that you are invisible to a specific buyer segment tells you exactly where the content and positioning gap exists. What we found is that brands that addressed segment-specific invisibility — by creating content that matches the vocabulary of each buyer segment — saw the fastest improvements in their AI visibility scores.

The pricing information problem in SaaS

What we found when auditing B2B SaaS brands is that pricing information is the most frequently wrong element in AI responses. AI models pull pricing from stale G2 listings, outdated comparison articles, and cached versions of pricing pages. In one audit, AI told a prospect a tool cost $30/user when the actual price was $10/user. In another, AI recommended a free tier that had been discontinued two years prior. These errors directly affect buyer perception and conversion at the point of purchase consideration.

Why some SaaS products dominate AI

The SaaS brands that consistently appear in AI recommendations share common characteristics: extensive third-party coverage on G2, Capterra, and industry publications; content that uses buyer vocabulary rather than internal product terminology; presence in analyst reports (Gartner, Forrester); and factual consistency across all indexed sources. None of these are about product quality directly — they are about information architecture and discoverability.

Last updated: April 2026